DevOps: Dockerize our Python app, set up CI/CD, Kubernetes (k3s) and monitoring

Our backend is Python/FastAPI with PostgreSQL and a RabbitMQ running on a single Ubuntu Server VM on AWS EC2. Right now everything is manual: we SSH into the server, pull code and restart processes by hand. Deployments break regularly, there’s no rollback, and we only learn about downtime from customers. We want a clean reproducible setup that our two developers can maintain after you leave. Scope of work: 1. Containerization - Multi-stage Dockerfiles for the API and the worker (slim or distroless images) - docker-compose for local development 2. Kubernetes - Single-node k3s cluster on our EC2 instance - Helm chart for the app: Deployments, Services, ConfigMaps - Liveness, readiness and startup probes, resource requests and limits - Ingress with TLS via cert-manager - Rolling updates with rollback 3. CI/CD (GitHub Actions) - Run tests and linters on every pull request - Build and version Docker images - Auto-deploy to staging, gated (manual approval) deploy to production 4. Infrastructure as Code - Terraform for the EC2 instance, VPC and security groups - Ansible playbook for server provisioning and Linux hardening 5. Monitoring & logging - Prometheus + Grafana dashboards: API latency, error rate, CPU/memory, queue length - Loki for centralized logs - Alerts for downtime and error spikes 6. Secrets & backups - Move secrets out of .env files into HashiCorp Vault - Automated PostgreSQL backups to S3 with a tested restore procedure 7. Handover - README and a short runbook: how to deploy, roll back, restore and read the dashboards Must have: Docker, Kubernetes, Helm, GitHub Actions, Terraform, Linux (Ubuntu Server), Prometheus/Grafana Nice to have: Ansible, Vault, Loki, experience with Python/FastAPI backends